A NOVEL ARTIFICIAL IMMUNE NETWORK ALGORITHM

摘要：

A novel artificial immune network (AINet) algorithm is proposed in this paper. In the algorithm, a new method is introduced to confirm searching radius, which is based on the age-the generations when the cell exists in the network.Moreover, a novel preserving method is proposed to avoid the instability and degradation of the optimal results. These two measures not only increase the local searching efficiency of the AINet algorithm, but also improve the diversity of the network.As a result, the computational complexity is significantly decreased. Simulation experiments show that, compared with the canonical AINet, the proposed algorithm enhances the correctness by 6％ and the speed by 34％ at least.

关键词： Artificial Immune Algorithm Immune Network Function Optimization

作者: XIANG TIAN HAI-DONG YANG FEI-QI DENG

作者单位: College of Automation Science & Engineering, South China Univ.of Tech., Guangzhou 510640, China;Scho College of Automation Science & Engineering, South China Univ.of Tech., Guangzhou 510640, China

会议类型: 国际会议

会议名称: 2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

会议地点: 大连

会议语种:英文

页码: 2159-2165

在线出版日期: 2006-08-13（万方平台首次上网日期，不代表论文的发表时间）